Ministral 3
modelEdge-optimized model family (3B, 8B, 14B) created via Cascade Distillation from Mistral Medium 3.1. 3 variants each (base, instruct, reasoning) = 9 models. 256K context. Image understanding. 14B closely matches Mistral Small 3.1 while 40%+ smaller. Apache 2.0.
Model Details
Architecture DENSE
Variants
| Name | Parameters | Notes |
|---|---|---|
| Ministral 3 3B | 3B | — |
| Ministral 3 8B | 8B | — |
| Ministral 3 14B | 14B | — |
Paper
arXiv: 2601.08584